A novel assay of neuronal cell adhesion

Vol. 63, No. 3, 1975

BIOCHEMICALAND BIOPHYSICAL RESEARCH COMMUNICATIONS

A NOVEL ASSAY OF NEURONAL CELL ADHESION

David I. Gottlieb and Luis Glaser Department of Biological Chemistry Division of Biology and Biomedical Sciences Washington University St. Louis, Missouri 63110

Received

February

19,1975

Intact cells can be made to adhere to a glass surface which is derivitized with y-aminopropyl-triethoxysilane, and then treated with glutaraldehyde. Such cells remain fully viable and a monolayer of such cells can be formed within 1 hour after the cells are removed from an embryo. The rate of binding of radioactively labeled single cells to such a monolayer can be used as a measure of intercellular adhesion (Walther, Ohman and Roseman, Proc. Natl. Acad. Sci. USA 70, 1569-1573 (1973). The lack of toxicity of this surface is shown by the fact that it supports normal growth of C-6 rat glial cells.

Recent work in this (1,2) and other laboratories

(3) has shown that embry-

onic brain cells have surface determinants which allow for selective intercellular adhesion.

Using the inhibition of cell aggregation by plasma membranes

we have defined both temporal and spatial changes in cell surface specificity. Anatomical and physiological studies show that the major brain regions are exceedingly complex with respect to cellular architecture.

The optic tectum and

the retina in which adhesive specificity has been most thoroughly studied consist of five and three major cellular layers respectively,

the cells of which

are distinguishable by numerous anatomical, physiological and biochemical criteria.

In attempting to investigate the cell surface specificity of these

cell types an assay capable of measuring the cell surface specificity of a small number of cells is essential. Walther et al. (4) have described an assay for cell adhesion based on the adhesion of radioactive cells to a monolayer of homologous or heterologous cells.

A disadvantage of this assay for studies of specificity in the devel-

Copyright © 19 ~ by Academic Press, ~ c . All r~hts o f r~roduction rese~ed.

815

Vol. 63, No. 3, 1975

BIOCHEMICALAND BIOPHYSICAL RESEARCH COMMUNICATIONS

oping central nervous system is the requirement that the monolayer be established in culture for 48 hours to have adequate mechanical strength to be useful in the assay, and as we have shown (2) considerable changes in cell surface specificity can take place in a 48 hour period. We have developed a variant of this assay in which cells can be made to adhere instantly to glass derivitized with the protein linking reagent glutaraldehyde activated y-aminopropyl-triethoxysilane.

A monolayer can be formed

within 1 hour after the cells are removed from the embryo, and the cells remain fully viable. MATERIALS AND METHODS Standard scintillation vials were cleaned in nitric acid.

2 ml of a 10%

solution of y-aminopropyl-triethoxysilane in toluene was added to each vial, the vials were covered with a glass marble and heated on a hot plate for 1 hour under gentle reflux.

The vials were allowed to cool and were thoroughly rinsed

with fresh toluene and allowed to air dry (5). glutaraldehyde

(Sigma) was pippeted into the bottom of the vial and left for 1

hour at room temperature. and air dried

A 1% aqueous solution of

(5).

The vials were thoroughly washed with distilled water

Single cell suspension of neuronal cells were prepared as

described previously

(i).

The cells were suspended in Dulbecco's modified Eagle's

medium (6), buffered to pH 7.3 with HEPES, and kept at 0 °.

107 Cells in 1 ml

of medium were pippeted into 2 ml of cold medium in a derivitized vial.

The

vials were immediately centrifuged in a Sorvall GLC-2 centrifuge with swinging bucket rotor at 162 x g for 5 min.

It is important that all solutions re-

main cool until the centrifugation step to avoid clumping of the cells.

After

centrifugation the vials were placed in a 37 ° bath for i0 min to stabilize the monolayers.

The medium was then decanted and replaced with 1 ml of calcium

and magnesium-free Hanks solution containing 5 mg/ml bovine serum albumin (Sigma)

(CMF-A).

The monolayers were then used in adhesion assays as described

in the legend to Figure 2. RESULTS Viability of cells in monolayer and stability of the monolayer. 816

It is

Vol. 63, No. 3, 1975

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

!7I lOs

1041

I

~

20

Figure i.

I

[

40

I

I

~

I

60 80 HOURS

I

1

100

I

i

i

120

Growth of C-6 cells in derivitized and untreated vials.

8 x 104 Cells were inoculated into derivitized or untreated vials containing 2 ml of Higuchi's (7) medium + 10% horse serum. Vials were incubated at 37°C under seal. Cells were removed from the bottom of the vials for counting by sixty minutes digestion with 0.15% trypsin in PBS (phosphate buffered saline pH 7.3) and counted in a haeomocytameter. ; $, cells grown in derivitized vials; 0 O, cells grown in untreated vials.

crucial that cells not be damaged during the construction of the monolayers. Three lines of evidence show that monolayers to the cells.

can be formed without gross injury

First, cells retained the ability to exclude trypan blue for

up to three hours after being "plated" on derivitized glass. cells on monolayers incorporate

Secondly,

the

[3H]leucine into protein at the same rate as

cells in suspension cultures over the first several hours after removal from the brain.

Finally, we have succeeded in growing the C-6 rat glial cell line on

derivitized glass and have shown that this surface is more favorable for growth than untreated glass

(Fig. 1).

This latter control is not strictly comparable

to the formation of embryonic cell monolayers

since the C-6 cells grow in the

presence of serum while the monolayers are plated in its absence. ever, show that the treated glass is not toxic to the cells.

It does, how-

The stability of

the monolayers was estimated quantitatively by counting the numbers of cells appearing in the medium after incubation on the shaker bath.

After 45 minutes

incubation in CMF-A less than 0.3% of the cells were recoverable

817

in the medium.

Vol. 63, No. 3, 1975

BIOCHEMICALAND BIOPHYSICAL RESEARCH COMMUNICATIONS

Evidence that added cells adhere to the monolayer.

Since the 32p labeled

cells leak about 20% of their label into the medium, it is important to show that this material does not accumulate in the monolayer. were performed by preincubating

Control experiments

labeled 9 day tectum cells in CMF-A for 45 minutes,

spinning out the cells and incubating the supernatant with fresh 9 day tectum monolayers.

The monolayers accumulated

compared to monolayers

less than 1% the number of counts when

incubated with an equivalent volume of cell suspension,

and soluble transfer is therefore not responsible for a significant portion of the labeling of monolayers. Unlike the monolayers studied by Walther et al (4) ours are fully confluent, with no visible gaps.

Therefore,

vial must be stuck to other cells.

cells which stick to the bottom of the

Over 85% of the bound label in one experi-

ment was removed by gently scraping the monolayer off the vial bottom while taking pains to avoid disturbing any cells which might be clinging to the sides of the vial.

Therefore, no more than 15% of the label in a typical experiment

could be found on the sides of the vial.

In fact this figure is probably con-

siderably lower since fragments of the monolayer remained on the bottom of the scraped vials. In spite of these experiments

an additional control was done to show that

adhering counts represent cell-cell adhesions rather than cells sticking to glass.

Fig. 2 shows the effects of temperature on 9 day tectal cells sticking

to homologous monolayers and to derivitized vials without a cell monolayer.

The

experiment shows that sticking to the monolayer is almost completely dependent on temperature,

whereas binding to the derivitized vials has virtually no tem-

perature dependence.

This supports the conclusion that most of the binding in

our experiments is cell-cell binding. Patterns of specificity between embryonic brain cells.

The optic tecta

and telencephala of the 9 day chick embryo were chosen for study because of their large size and ease of dissection.

When tectal and telencephalic

were tested for their ability to bind to tectal monolayers,

818

single cells

a surprising re-

Vol. 63, No. 3, 1975

BIOCHEMICALAND BIOPHYSICAL RESEARCH COMMUNICATIONS

30

37°C

~o ~

0oc

lO

i

20

40 MINUTES

i

i

20

i

40

Figure 2. Temperature dependence of embryonic brain cell adhesion to monolayers and derivitized glass. Embryonic brain cells were labelled with 32p by incubating triturated organs in 5 ml of DulbeccOs m ~ i f i e d Eagle's medium (6) + 10% chicken serum with 500 NCi of carrier free HR PO, for 3 hours at 37°C. The tissue was then transfered 4. to the same medium without Isotope and incubated for an additional hour. The tissue was then used to prepare single cells by standard methods of this laboratory (1,2). For the adhesion assay, each vial w a s inoculated with 1 x 105 labelled cells (.17 cpm/cell). Vials were placed in gyratory water baths set at 0°C and 37°C and agitated at 65 rpm for appropriate times. Cell suspensions were then removed by gentle aspiration and the vials carefully washed twice with 5 ml of chilled PBS. 1 ml of 1% Triton XIO0 was added to the vials to dissolve cells. 3a70 Counting fluid (RPI Corporation, Elk Grove, Illinois) was added and the vials counted in a scintillation counter. Left panel.

Binding to monolayers,

Right panel. Binding to derivitized

g ~, 37°C 0----O, 0°C

glass, • O

sult was obtained, namely that the telencephalic than the tectal cells.

The converse experiment

pression of a universal stickiness

shows that this is not an excells

are tested for their ability to stick

the tectal cells now show the greater affinity.

This reciprocal affinity is highly reproducible. heterotypic binding always exceeded homotypic monolayers.

cells showed greater binding

on the part of the telencephalon

because when the same cell suspensions to telencephalon monolayers,

~, 370C O, 0°C

In five such experiments

to both tectal and telencephalic

Part of the difference appears to be a result of the relative

rates of binding as the early time points always indicate that heterotypic binding is favored over homotypic.

Most experiments

show that the extent of

binding is greater for the heterotypic pairs although this aspect of the data

819

Vol. 63, No. 3, 1975

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Tectum Monolayer 3C

8 2C

Tec,um Monoloyer

7Y i

;

20

Fisure 3.

Telencep~alon Monolayer

p

i

40

i

i

i

i

i

20 40 MINUTES

l

20

i

,

40

Specificity of cell adhesion.

4 9-Day tect~m cells (7.7 x i0 cells, .16 cpm/cell), H ; 9 day telencephalon cells (~.0 x i0 ~ cells, .15 epm/cell), 0 O; and Chinese hamster ovary cells (6 x l0 T cells, .53 cpm/cell), 0 [], were inoculated into vials containing appropriate monolayers. All experiments were conducted at 37°C with the platform rotating at 65 rpm. Kinetics of binding were determined as in Figure 2.

shows more variability.

Chinese hamster ovary cells

(a totally heterologous

cell type) essentially does not bind to tectal monolayers. The binding of tectum and telencephalon cells to monolayers

gives the sur-

prising result that heterotypic binding is preferred to homotypic binding. Two conclusions complementary typic pair.

can be drawn from these data.

type in that the heterotypic

i) The specificity

is a

case is preferred over either homo-

2) The fact that the same result is obtained in reciprocal pairs

is an indication that being placed in a monolayer does not damage the cell's specific recognition apparatus. In this assay we measured the adhesion of 105 free cells to 107 cells in a confluent monolayer.

It is probable that cells in the monolayers are alter-

ing one another's surface in complicated ways. sites may be masked by "self" interactions cell-cell interactions

For example, homotypic binding

in the monolayers.

The effect of

in the monolayers on cell surface specificity is under

investigation. The derivitized glass products provide useful new methods for the study of cell adhesion and a potentially useful new surface for cell growth studies.

This work has been supported by Grants NIH GM 18405 and NSF BMS75-22638 and NIH F22-NS 166.

820

Vol. 63, No. 3, 1975

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

REFERENCES i. 2. 3. 4. 5. 6. 7.

Merrell, R. and L. Glaser (1973) PNAS 70:2794-2798. Gottlieb, D.I., Merrell~ R. and L. Glaser (1974) PNAS 71:1800-1802. Barbera, A.J., Marchase, R.B. and S. Roth (1973) PNAS 70:2482-2486. Walther, B.T., Ohman, R. and S. Roseman (1973) PNAS 70:1569-1573. Robinson, P.J., Dunnill, P. and M.D. Lilly (1971) Biochim. Biophys° Acta 242:659-661. Dulbecco, R. and G. Freeman (1959) Virology 8:396. Higuchi, K. (1970) J. Cell Physiol. 75:65-72.

821